BRIEF
POLICY
Issue 2017/08
April
2017
Competition Policy and
E-Commerce after the EU
Commission Sector Inquiry:
What Comes Next?
Marco Botta & Maria Luisa Stasi, ENTraNCE
Abstract
On 3 March 2017, the Robert Schuman Centre for Advanced Studies (RSCAS) and Assonime (Association of Italian Joint Stock Companies) jointly organised a workshop at the European University Institute (EUI) in Florence. The workshop focussed on the role of EU competition policy in e-commerce, in the light of the recent European Commission Sector Inquiry relating to this field.
The event was opened with a keynote speech delivered by Cecilio Madero Villarejo, Deputy Director General for Antitrust of the DG Competition of the European Commission. The workshop was divided into three panels, which dealt respectively with (i) economic insights from the EU Commission Sector Inquiry in relation to the e-commerce of goods; ii) the enforcement approach followed by National Competition Authorities (NCAs) in e-commerce; and iii) e-commerce in digital content.
The event gathered together different stakeholders, including representatives from NCAs, academia, industry, law firms and economic consultancies. The diversity of views ensured a lively debate. This Policy Brief summarises the main points raised during the discussion and seeks to stimulate further debate.
Background: EU Commission Sector
Inquiry in relation to e-commerce
In May 2015, the DG Competition of the European Commission launched a Sector Inquiry in relation to e-commerce as part of the Digital Single Market (DSM) strategy. In September 2016, DG Competition published the Preliminary Report resulting from its Sector Inquiry.1
In particular, the Preliminary Report focussed on the online sale of goods and digital content. The European Commission is expected to publish the Final Report on the Sector Inquiry in the course of 2017.
During the Sector Inquiry, the EU Commission collected information from over 1800 companies, including manufacturers, retailers and copyright owners of digital content. In addition, the EU Commission scrutinized over 8000 distribution agreements that included clauses affecting e-commerce of goods and digital contents. Finally, 65 firms participated in the public consultation that followed the publication of the Preliminary Report in Autumn 2016. The Sector Inquiry thus provided the EU Commission with an overview of the competition dynamics in the e-commerce industry in Europe.
The Preliminary Report pointed out that e-commerce enhances online price transparency thus strenghtening price competition and increasing the consumers’ choice. In particular, consumers can easily compare products and prices via price comparison tools, decreasing their search costs and allowing consumers to switch from one retailer to another. Nevertheless, the Preliminary Report also identified a number of trends that may raise anti-competitive concerns. First of all, the Preliminary Report pointed out that most retailers track online prices of their competitors and a majority of them use automatic software programmes to adjust their prices in view of those of competitors. In particular:
1. 53% track online prices of their competitors;
2. 67% of retailers use automatic software for that purpose;
3. 78% of retailers use software to track prices and to adjust their own prices to those of their competitors;
1. The Preliminary Report is available at: http://ec.europa.eu/ competition/antitrust/sector_inquiries_e_commerce.html (last visited 23.3.2017).
Secondly, manufacturers have adopted a number of practices in order to better control the online distribution of their products and to reinforce competition on other parameters rather than prices, such as brand image and quality. In fact, while retailers compete primarily on price, manufactures compete mainly on quality and brand image.
The Preliminary Report identified three main reactions by the manufacturers vis a vis the growing e-commerce: 1. Selective distribution systems in which the products
can only be sold by pre-selected authorized sellers are used more widely;
2. Manufacturers increasingly sell their products online directly to consumers, thereby directly competing with their own retailers;
3. Manufacturers resort to vertical restraints.
Among the typical restraints affecting e-commerce that manufacturers include in vertical agreements the Preliminary Report pointed out that:
1. 42% of the retailers face some form of price recommendation or price restriction from manufacturers; while pure recommendations are not illegal, some of them amount to Resale Price Maintenance (RPM);
2. 18% of the retailers are contractually restricted from selling on online marketplaces (i.e. platform bans); 3. 9% of the retailers are contractually restricted from
submitting offers to price comparison websites; 4. 11% of the retailers have reported to the EU
Commission that their suppliers impose contractual restrictions on cross-border sales.
These types of contractual sales restrictions may, under certain circumstances, make online shopping more difficult and ultimately harm consumers, by preventing them from benefiting from greater choice and lower prices offered by e-commerce. On the other hand, in relation to the widely debated platform bans, the preliminary findings of the European Commission call for a nuanced approach which does not warrant a general prohibition of such bans (i.e. no hardcore restriction), but rather a case-by-case approach based on the effects of individual restrictions.
In relation to the e-commerce of digital contents, the Preliminary Report pointed out that one of the key determinants of competition in digital content markets is the availability of licences. The availability of rights for online transmission is largely determined by the scope of rights, as defined in the licensing agreements between right holders and digital content providers. The European Commission will therefore need to assess whether certain licensing practices in digital content markets restrict competition, taking into account the legal and economic context and the characteristics of the specific product and geographic markets, and whether competition law enforcement is necessary in order to ensure effective competition.
E-commerce in goods: economic insights
from the Sector Inquiry
The economists participating in the first panel of the workshop pointed out that the competitive problems identified by the EU Commission in the context of the Sector Inquiry are well-known issues in the economics of vertical restraints. In particular, clauses in vertical agreements that limit the ability of the distributor to sell goods online/ or to set the price may generate foreclosure concerns and favour the cartelisation at the downstream level. On the other hand, such clauses may also have pro-competitive effects and be justified for legitimate reasons. For instance, geo-blocking allows price discrimination among different customers who are based in different countries. Such discrimination may reflect the differences in demand conditions in different EU member states, based on the consumers’ purchasing power. Price discrimination caused by geo-blocking may therefore increase consumers’ welfare, rather than harming it. Besides their ambiguous welfare effect, vertical restraints in e-commerce may be justified for reasons of ‘incomplete contracting’: selected retailers provide to the final consumers a number of additional services that are not originally included in the distribution agreement with the manufacturer (e.g., presenting the product to the final consumers; assisting consumers in the case of the repair of the product). By limiting intra-brand competition, vertical restraints safeguard the incentive of the distributor to provide such additional services, for which the retailer does not receive an express remuneration from the manufacturer. In the context of e-commerce, vertical restraints may safeguard bricks and
mortar shops that usually provide such additional services to final consumers; the latter would otherwise disappear due to the free-riding enjoyed by online retailers.
Besides the prevention of free riding and remuneration for incomplete contracting, vertical restraints are required to safeguard the brand image of the product. In the context of online commerce, in fact, the traditional trademark protection is not sufficient to safeguard the brand image: firms rely on specific colours, designs and product presentations that allow the consumers to identify the brand. , but such details are easy to replicate by others in online sales channels. In other words, it is easy to affect brand image if online channels are not used properly. Manufacturers are keen to ensure that the brand is presented in the same manner in both bricks and mortar shops and in online marketplaces. As a consequence, manufacturers usually impose clauses in vertical distribution agreements that limit the ability of retailers to sell the goods on web-sites that do not safeguard the appropriate brand image.
Another quite diffuse vertical restraint, which is linked to the brand and/or the trademark’s exploitation and safeguarding, concerns the restriction on retailers to bid for the trademarks of certain manufacturers in order to get a preferential listing by the search engines of paid referencing services. In this type of auction market, the supplier usually has strong market power, while there is fierce competition on the demand side, often with inefficient outcomes. Precluding preferential listing is a discriminatory behaviour that directly affects traffic and sales and raises rivals’ costs. The challenge here is to evaluate whether the restraint can be justified for trademark protection reasons, or for other plausible economic justifications, otherwise it may be considered as a by-object restriction.
The participants agreed that the increasing reliance by manufacturers on vertical restraints in selective distribution agreements represents a reaction to the growth of e-commerce. In particular, due to the potential efficiencies and legitimate reasons mentioned above, vertical restraints in e-commerce should be subject to case-by-case analysis, rather than being considered to be hard-core violations. This is the case, in particular, for vertical restraints that have recently emerged in the e-commerce world and which did not previously exist in traditional selective distribution agreements (e.g., a ban on the online marketplace and price comparison
web-The Adidas and Cite cases2 were mentioned as examples
of this uncertainty.
It was suggested that, given the difference in approach across the EU, an analysis of the economic impact of both the use of the contested practice and its total prohibition could help to provide guidance on future action. In particular, the effects should be considered EU-wide, rather than in each country. Moreover, these effects are quite different for big companies and for SMEs, and this is not only a German phenomenon. This being the scenario, participants wondered whether, and how, the NCA’s intervention changes it. In other words, they wondered about the counter-factual, which so far does not appear to have been taken into due consideration. Perhaps, a generalised intervention could help bring prices down at an EU level. There was nevertheless a consensus on the fact that, while reasoning about prohibiting clauses, the economic and business reasons for such differences should be taken into account, together with the legal ones.
Recently, the line between price recommendation and minimum price fixing has been difficult to draw. Frequently, price recommendation mechanisms imply the use of software and algorithms, and this may affect the assessment of conduct. Here, some Member States, like Germany, are using a strict approach: in the CIBA case3, for instance, the German NCA concluded that the
mechanisms for systematically monitoring competitors’ prices would lead to an infringement of the competition rules. In this respect, participants debated on the concept of interdependent action, its boundaries and its role in the assessment of the conducts at stake, all elements that have so far neither been clarified by the European Commission, nor by the NCAs.
Uncertainties and unresolved issues are not new in the field of vertical restraints; but in the online dimension their impact is wider. In more general terms, there was consensus around the fact that a better understanding of vertical restraints, and not just of specific clauses, could be useful in assessing when and how to intervene. In particular, participants expressed the need for a greater understanding of the likelihood, size and nature of
free-2. http://www.bundeskartellamt.de/SharedDocs/Meldung/ E N/ P re s s em it tei lu ngen /2 014 /02 _ 07_ 2 014 _ ad id a s . html?nn=3591568 (last access 3.4.2017).
3. http://www.bundeskartellamt.de/SharedDocs/Meldung/EN/ Pressemitteilungen/2009/25_09_2009_Ciba-Vision.
sites). In addition, in the context of e-commerce, the European Commission should re-consider the traditional
per se approach that is followed in relation to a number of
vertical restraints (e.g., RPM and restrictions on passive sales), where a number of reasons call for a case-by-case, rather than a per se, analysis.
The enforcement approach followed by
National Competition Authorities
While opening this session, it was noted that there are major differences across Member States in relation to the assessment under Article 101 TFEU of many of the critical clauses that are typical of the e-commerce sector and that restrict competition, and also in relation to the enforcement strategies followed by NCAs. On the one hand, Regulation 1/2003 calls for rules to be applied uniformly by national enforcers; however, the Sector Inquiry shows the opposite in the e-commerce sector. Regulation 1/2003 also sets the grounds for coordinated enforcement by means of the European Competition Network. Again, this does not appear to be the case in the sector being investigated. Finally, Regulation 1/2003 gives the European Commission the role of focussing on the major cases, and the cross-countries ones; nor, here, has the Sector Inquiry confirmed this circumstance. Concerning geo-blocking, so far, very little enforcement has taken place at the national level, while the European Commission is currently running two investigations. Participants noted that one of the reasons may be that geo-blocking has a fundamental role to play in the full deployment of the Digital Single Market (DSM), and thus this may not be an issue with which the NCAs should deal.
MFNs clauses, on the contrary, show a more uncertain state in the law. In this field, NCAs have been quite active, especially with regard to “narrow” clauses. However, if, on the one hand, this experiment leads to mutual learning and allows for a better observation of the effects of different approaches, on the other, doubts were cast concerning the outcomes, in terms of legal certainty. Another point in which the EU competition law appears to be quite uncertain is the ban on platforms, which is subject to different guidelines in various Member States, and to different use by the national business communities. It remains unclear whether the VBER applies, or whether the ban has to be considered to be a hard core restriction.
riding, in order to establish under what circumstances the latter can justify vertical restraints.
To conclude, it was stressed that, currently, there is relatively little coordination of public enforcement. Rather, NCAs seem to compete. On the other side, though, it was pointed out that the European Commission seems to adopt a ‘wait and see’ approach; and participants wondered about the effectiveness of this choice, because the sector is fast moving, and therefore the lack of timely intervention may be difficult for businesses to manage. While discussing enforcement strategies, it was suggested that NCAs should concentrate on the facts and try to show evidence of any reduction in economic welfare, which is what matters to the consumers. Another strategy could be to go beyond simple command and control; in fact, infringement decisions are not the only tool at the NCAs’ disposal. Commitments can also be used, although they usually result in limited judicial review and have little value in relation to precedent. Attendees argued that it might be more efficient to better explain, in a prohibition decision, how the relevant rules should be interpreted and applied. Overall, the NCA’s legitimacy would be enhanced by the latter’s clarity.
Additionally, the adoption of positive decisions should be taken into due consideration, as well as the possibility of providing better guidance through sector inquiries or guidelines. Finally, the possibility to build a profitable dialogue with the business community should not be excluded, in order to better understand the market and business dynamics in a very fast-changing environment. In the light of the latter, a regular ex post review of positive and negative decisions could bring many advantages. It was also noted that, so far, NCAs have rarely used consumer protection tools in the e-commerce sector, but they might consider doing so. However, the risks of creating disparities in the level of protection, grounded on the disparities of the NCAs’ powers relative to consumer protection in the different Member States, should also be carefully considered when reflecting on this option. As for the tasks being shared between the Commission and the NCAs, attendees raised a number of points, both to favour the former over the latter, and vice-versa. On the one hand, the European Commission should move first, because it can pick the best cases; it can issue positive decisions; it can create precedents to follow. On the other
hand, the NCAs should act first in order to allow for experiment, which, in turn, creates the conditions for mutual learning; moreover, NCAs might take into due account the economic differences among the national markets; finally, NCAs can also adopt cross-jurisdictional decisions, although they appear extremely reluctant to do so. This is a key point, since e-commerce cases can easily show cross-border effects, as, for example, when the incriminated clause damages the consumers of other countries, and this is a de facto novelty (with which NCAs have to cope. In this context, the role of the European Competition Network (ECN) was also discussed: while some of the participants held the view that Network had played a limited role in the OTAs case, others emphasized that thanks to ECN an acceptable level of coordination had been achieved, ensuring a convergent outcome in the majority of the EU member states.
The debate was enriched by the discussion of national experiences. Amongst others, it was noted that the
Bundeskartellamt seems to act as a leader, and seemed
to have an approach that is strongly oriented towards consumer protection. The CMA also appears to be very active and inclined to experiment, as it uses different tools to achieve results; moreover, its assessment is made on a case-by-case basis, rather than on per se rules. The AGCM, on its part, seems to combine enforcement and advocacy4 and to consider consumers’ empowerment
to be an effective solution. One might argue, though, that this solution is extremely sensitive to the level of consumers’ awareness and e-literacy.
E-commerce in digital content
The last panel of the day concentrated on two main issues, both identified as areas for action, including legislative action, by the European Commission: unjustified geo-blocking, and the need to modernise copyright rules. When dealing with digital content, the key determinant of competition is the availability of licenses from the holders of copyrights. However, contractual restrictions have become the norm, and many of them grant exclusivity, even those contractual restrictions that are long term (20+ years). On the other hand, geo-blocking is sometimes realised without the use of contractual clauses, but, rather, by means of unilateral decisions. Although exclusivity and geo-blocking are not used with
the same frequencies in all Member States, they create many concerns in terms of market partitioning and barriers to entry.
In general terms, participants recalled that freedom to choose one’s trading partner remains the basic principle, and that restrictions are allowed under national copyright rules. We should thus be out of the scope of per se rules, and we need to apply a case-by-case approach.
One element to take into account is that, when it comes to digital content, there is a potential friction between the DSM objective and copyright rules, because the latter grant exclusivity on a territorial basis and allow for the use of territorial licenses. Participants noted that, currently, there is a strong political will towards the realisation of the DSM, and this may have a deep impact on where the balance between free movement of services and competition law principles, on the one hand, and the protection of IPRs, on the other, should be struck. Some also wondered about the risks of over-enforcement. Finally, the comparison was made with the recent EU experience in relation to roaming charges for mobile calls.
Furthermore, it was noted that audio-visual markets are still national in scope because of cultural differences. According to the EU rules, exclusive territorial restrictions are permitted, while absolute restrictions are not. The key point here is thus where the line should be drawn between passive and active sales in the online world. It was mentioned that, while doing this, the European Commission might change the behaviour of an entire industry.
When dealing with restrictions related to the distribution and sales of digital content, the uneasy relationship between competition and IP rules comes into play. Articles 101 and 102 of TFEU, Article 20(2) of the Service Directive (which prohibits discrimination on the basis of nationality or place of residence) and the VBER, are all applicable, depending on the circumstances. However, where the restriction falls within the ‘specific subject matter’ of the IP right, if it is considered reasonable and proportionate, then it is permitted under sector specific rules.
Among the sector specific rules, the principle of exhaustion was mentioned, which aims to mitigate the principle of territoriality and to stimulate cross-border exploitation of copyrighted content. The key point is
that exhaustion does not arise in the case of online services5. Territorial licensing of intangible copyrighted
works and services is thus lawful, and parallel imports can be prevented by relying on the territoriality of copyright. As there are no borders on the Internet, the only way to establish territoriality for digital content is through geo-blocking, which the European Commission has recently declared a restriction by object in the Sky-UK case.6 However, the case was criticised, suggesting that
the Commission’s analysis of market definition, market power and the justifications put forward by the parties was not sufficiently performed.
In addition, participants raised the point that, due to the importance of the sunk costs and the short life of products in both the music and the audio-visual industries, the possibility to discriminate between Member States through prices can be essential to ensure the viability of business models and to preserve the dynamic incentives to invest in the production of creative content. In other words, price discrimination can increase welfare. When geo-blocking supports this price discrimination, it may thus be looked at in a different way, and not as a restriction by object.
On the contrary, it might even be argued that, in these specific cases, to pursue the DSM objective by prohibiting any form of price discrimination could actually damage consumer welfare. To conclude, there appeared to be consensus around the need to look with more attention at the concrete effects of single discrimination, which are difficult to assess ex-ante, rather than adopting a by object approach.
It was noted that, so far, the European Commission has given almost no guidance about how to deal with the issues mentioned above, while the EU Courts have. The principle laid down in Murphy7, according to which
the protection of copyright should not be equated with exclusivity, and much less with territorial exclusivity, was recalled, and participants agreed that it is applicable beyond its own framework. Embracing the opposite view
5. Article 4, InfoSoc Directive.
Directive 2001/29/EC of the European Parliament and of the Council of 22 May 2001 on the harmonisation of certain aspects of copyright and related rights in the information society. OJ L-167/10-19, 22.6.2001
6. http://europa.eu/rapid/press-release_IP-15-5432_en.htm (last access 3.4.2017).
7. Case C-403/08, Football Association Premier League Ltd and
would produce a result that is incompatible with the internal market aims and EU law by partitioning national markets. However, there was consensus on the need for very modern rules to fit the digital age. Extending the logic of the Murphy ruling to the online transmission of content is possible, in theory; however, participants wondered whether, in practice, this can happen without harmonising national copyright regimes, at least in order to extend the ‘country of origin’ principle to online transmission.
A number of additional Court cases providing precious indications were recalled and debated. For instance, in Coditel the Court established that the exhaustion principle might articulate differently for broadcasting.8
However, there was consensus that various issues still remain unsolved. By way of example, participants discussed the possibility of applying Article 101(3) on the basis of increasing incentives to invest in audio-visual content, or the encouragement of translations, and of other similar non-competition objectives. Furthermore, they also wondered whether, in light of the economic and legal context of the specific case, reverting to competition tools might not always be the best solution.
In conclusion, the various on-going legislative initiatives were mentioned, and their respective capacity to deal with current challenges and frictions was debated.
8. Case C-324/07, Coditel Brabant SA v Commune d’Uccle and
Robert Schuman Centre for Advanced Studies
The Robert Schuman Centre for Advanced Studies (RSCAS), created in 1992 and directed by Professor Brigid Laffan, aims to develop inter-disciplinary and comparative research on the major issues facing the process of European integration, Euro-pean societies and Europe’s place in 21st century global politics. The Centre is home to a large post-doctoral programme and
hosts major research programmes, projects and data sets, in addition to a range of working groups and ad hoc initiatives. The research agenda is organised around a set of core themes and is continuously evolving, reflecting the changing agenda of European integration, the expanding membership of the European Union, developments in Europe’s neighbourhood and the wider world.
ENTraNCE
European Networking and Training for National Competition Enforcers-is a project of the Robert Schuman Centre for Ad-vanced Studies at the European University Institute, which focuses on competition law and economics. ENTraNCE acts as a hub where European and international competition enforcers and other stakeholders can exchange ideas, share best-practices, debate emerging policy issues and enhance their networks. In addition, since 2011, the Robert Schuman Centre for Advanced Studies organises a training for national judges in competition law and economics co-financed by DG Competition of the European Commission - ENTraNCE for Judges.
Content © Marco Botta, Maria Luisa Stasi. © European University Institute, 2017
QM-AX-17-008-EN-N
doi:10.2870/64044 ISBN:978-92-9084-469-3 ISSN:2467-4540
ENTraNCE
Robert Schuman Centre for Advanced Studies European University Institute Via Boccaccio, 121
50133 Florence Italy
Contact:
